Summary: Current experimental vaccines based on the HIV envelope glycoprotein (Env) have failed to produce broadly neutralizing antibody responses, despite the fact that HIV-infected individuals can generate such antibodies. This discrepancy suggests that current Env immunogens may not adequately mimic appropriate structures in Env. This project attempts to understand Env structure with respect to antibody binding and the generation of neutralizing antibodies. First, we are attempting to define the antigenic structure of various Env immunogens using monoclonal antibodies and other defined anti-Env sera. Second, we are making novel Env immunogens to test whether focusing the immune response on specific conserved, functional structures will generate broadly neutralizing antibodies (de Rosny et a. 2001 J. Virol. 75:8859-8863). The antibodies raised by the defined Env subdomains will further be used to map Env immunogens and regions involved in conformational changes required for entry. This approach has already demonstrated success in raising structure specific antibodies to Env that preferentially react with different conformations of Env as it undergoes changes during virus entry.